Reader Response: Association of Position Played and Career Duration and Chronic Traumatic Encephalopathy at Autopsy in Elite Football and Hockey Players

摘要

Schwab et al. 1did not find a statistically significant association between position played and career duration and autopsy-confirmed chronic traumatic encephalopathy (CTE) in elite football and ice hockey players. The study's limitations associated with statistical power, measurement error, and sample selection hinder interpretation of the presented data. The authors state that a sample size of about 14 is required "to detect a 50% difference in CTE diagnoses between groups with 80% power." Player position is unlikely to account for a 50% difference in CTE status based on work by Mez et al., which indicates no effect for player position in >250 former football players. Assuming that player position accounts for a much smaller effect in CTE status, such as 10%, a sample size of approximately 400 would be needed. In Mez et al., each additional year of football play accounted for a 30% increase in odds of CTE. Given the sample size and standard deviation reported for career duration in this article, the authors were only 80% powered to detect an effect 2.4 times larger and were only 55% powered to detect the effect presented in Mez et al. In addition, career duration was problematically measured. The authors state that "age of retirement was used as an indicator of overall career duration." However, elite athletes can retire at the same age, but have very different career lengths—specifically in football, where players begin tackling as early as age 4 years and as late as age 20 years. Furthermore, the authors equate football players with ice hockey players, which are 2 sports with very different repetitive head impact (RHI) exposure profiles. Given these concerns with statistical power and appropriate measurement of RHI exposure, we question the validity of the reported results.